Date of Award

Fall 2011

Document Type


Degree Name

Doctor of Philosophy in Biology - (Ph.D.)


Federated Department of Biological Sciences

First Advisor

Sanjay Tyagi

Second Advisor

Farzan Nadim

Third Advisor

Edward Michael Bonder

Fourth Advisor

Alexis J. Rodriguez

Fifth Advisor

Arun Raj


Gene expression is a stochastic process, with elements of randomness present in both transcription and translation. This stochasticity results in cell-to-cell variation in the amounts of gene products, mRNAs and proteins, and is observed in organisms ranging from bacteria and yeast to higher eukaryotes. Randomness in the activation and inactivation of a gene is the preliminary cause of this variation. At the level of proteins, these variations are buffered compared to levels of mRNA, due to the longer lifespan of proteins. Nevertheless, there is substantial variation observed at the level of proteins, resulting in phenotypic diversity among genetically identical cells.

In higher eukaryotes, sets of genes are often expressed in a coordinated manner, and function together in response to extracellular stimuli. If the expression of such genes is indeed stochastic, how can a given cell produce a coherent response? Additionally, during multi-subunit protein assembly, how does variation in levels of the component proteins affect their assembly and impact their function? Furthermore, how does this variation propagate in a gene regulatory pathway, when protein products of an upstream gene, or a pair of upstream genes, aids in the expression of downstream genes? Does variation in the expression of upstream genes affect the expression of downstream genes?

These questions are addressed using the serum-mediated induction of c-Fos and c- Jun as a model. c-Fos and c-Jun are transcription factors that together form heterodimers and induce the expression of downstream genes. With the aid of single-molecule fluorescence in situ hybridization for the detection of individual mRNA molecules, cell­to-cell variation in the expression of c-Fos and c-Jun mRNAs, and variations in the expression of mRNAs from a pair of downstream genes, collagenase and cox-2 were studied. Cell-to-cell variation in the number of c-Fos-c-Jun protein heterodimers in the nucleus was also studied. It was found that, although c-Fos and c-Jun mRNA expression is highly variable and not correlated, the number of the c-Fos-c-Jun protein hetrodimers did not vary as much from cell to cell. Despite relatively invariant heterodimer numbers, the downstream mRNAs, collagenase and cox-2, were expressed in a highly stochastic manner. These results suggest that, despite the buffering of variation in intermediate steps, the downstream steps in a gene regulatory pathway are noisy. These results are consistent with the view that noisy expression is an inherent property of the transcriptional machinery.

As a second project, where in the nucleus, and at what step during mRNA biogenesis, does mRNA splicing occur was explored. It is believed that splicing generally occurs co-transcriptionally at the gene locus. Introns are removed before the mRNA is released. However, during alternative splicing it is important that processing be delayed until all of the exons and introns involved in the splice choice are synthesized. Is processing just delayed briefly until the alternative splice sites are synthesized, or does alternative splicing involve the uncoupling of splicing from transcription, so that splicing occurs post-transcriptionally?

The intracellular distribution and dynamics of individual molecules of pre-mRNAs and their spliced products were imaged utilizing a set of synthetic reporter genes, as well as a classically well-studied alternatively spliced gene: Sex-lethal (Sxl) in Drosophila. The normally tight coupling between transcription and splicing was found to be broken in situations where an intron’s polypyrimidine tract is sequestered within a strong secondary structure. Furthermore, it was also found that, in the case of the alternative splicing of Sxl mRNA in female Drosophila cells, particular exon is removed from the transcript, due to the activity of the RNA binding protein Sxl, which binds to nearby introns, causing splicing in those regions to be uncoupled from transcription. This uncoupling occurs only on the perturbed introns, while the preceding introns are removed co-transcriptionally.

Included in

Biology Commons